diff --git a/src/c++/perf_analyzer/genai-perf/genai_perf/profile_data_parser/llm_profile_data_parser.py b/src/c++/perf_analyzer/genai-perf/genai_perf/profile_data_parser/llm_profile_data_parser.py
index d812858c5..577ed2986 100755
--- a/src/c++/perf_analyzer/genai-perf/genai_perf/profile_data_parser/llm_profile_data_parser.py
+++ b/src/c++/perf_analyzer/genai-perf/genai_perf/profile_data_parser/llm_profile_data_parser.py
@@ -220,6 +220,8 @@ def _get_openai_input_text(self, req_inputs: dict) -> str:
             return payload["prompt"]
         elif self._response_format == ResponseFormat.OPENAI_VISION:
             content = payload["messages"][0]["content"]
+            # When no images were included in the request input, the content
+            # is same as text-only chat completions format (e.g. string).
             if isinstance(content, str):
                 return content
             return " ".join(c["text"] for c in content if c["type"] == "text")